[Vorbis-dev] Vorbis and .NET

[illiminable]

----- Original Message ----- 
From: "Tor-Einar Jarnbjo" <Tor-Einar at Jarnbjo.de>
To: <vorbis-dev at xiph.org>
Sent: Sunday, October 03, 2004 4:09 PM
Subject: Re: [Vorbis-dev] Vorbis and .NET


> Nathan I. Sharfi wrote:
>
>> 1) Think that Ogg/Vorbis/FLAC/Theora/etc. rock the house
>> 2) Are willing and able to touch Managed C++ (or, in general, 
>> Microsoft-only
>>   technologies) with considerably less than a 10' pole.
>>
>> I suppose this is one more thing that's holding up adoption. Suggestions 
>> on
>> how to fix this problem, anyone?
>>
> I think a good start would be to admit that the Ogg formats really don't 
> rock the house compared to the default installation of media codecs on a

I'm not sure i'd go that far... vorbis and speex IMO are superior in most 
cases to what is part of the default install of codecs... FLAC until WM9 
too... but WMA9 Lossless beats FLAC almost every time at the moment in my 
experience (but in most cases not so substantially to be a major 
detterent)... theora... well the jury's still out on that at this stage, but 
it's still early days yet.


> Windows system, probably still having a market share of >>95% on home 
> computers. The next step is to admit that the "free" argument is rather 
> relative if you compare the license costs for the mentioned codecs with 
> the labour costs for developing a user friendly software solution for Ogg 
> based services like media on demand. If we also admit, that the multimedia 
> industry won't allow providers to distribute content without some sort of 
> DRM restrictions, I think we're getting close to reality.
>
> Sad. but true ...

I think the main thing for attracting the windows crowd is to understand 
that they are a totally different animal to their open-source/gnu/linux 
brethren. What works to attract developers/users to other platforms doesn't 
necessarily apply to windows. Customising the approach and the "message" to 
this audience is key IMHO.

I think some of the most important differences are (again my opinion and 
perception only) ;

1) The free-software culture is small on windows, the more common 
"free-like" models are freeware(no source), trialware, crippleware and 
personalware (non-commercial only is free).

2) In light of (1) one of the main reasons a lot of windows software is 
written is to make money.

3) Windows developers don't write for MS platforms because they think MS 
rocks, they do it mainly because it has the most users, and hence the most 
potential usership and/or revenue. MS continues to dominate, not just 
because they make a ton of money, but because they provide a platform to 
allow others to easily make money too, which in turn strengthens MS. 
Symbiotic (or at least psuedo-symbiotic) relationships are the key to most 
successful systems in nature and business.

4) Windows developers are probably less interested in philosophies. They are 
more interested in just giving the end-user what they want.

5) Windows users for the most part probably couldn't care less about 
philosophies, they just want stuff that does what they want, how they want 
it. Even if you explained to them the potential for world domination by a 
few powerful companies and how it may lead to the ultimate demise of quality 
content... they probably still don't care.

6) Windows users are less likely to get involved and report bugs 
(particularly if it involves effort and/or uncertainty of outcome).

7) Windows users in general prefer to send an email to a person, rather than 
join a mailing list just to report a bug/ask a question or use some bug 
tracking software that looks confusing , and i'd hazard a guess that many 
have never signed up to a mailing list in their life and even the simple 
instructions on how to sign up and/or the wait for approval of posts is more 
likely to have them hitting that delete key on your software.

8) Most Windows users will delete your software if they don't like it rather 
than tell you what they didn't like about it. They have plenty to choose 
from.

9) Windows users are fickle... ignore conventions and "their way of doing 
things" at your own risk.

10) Windows users in general are less computer savvy... if it's not easy to 
use, it won't be used. For at least 75% of users this means command lines 
are a no-no.

11) Windows developers are spoiled by MSDN (probably the most comprehensive 
set of publicly available software documentation) ... and massively active 
newsgroups, and just the sheer volume of other developers to help them. That 
is their expectation of developer support with the technologies they 
commonly use. And the last two are probably a product of the first in the 
sense that the barrier to entry is very low.

12) Many Windows users prefer a web "forum" layout than email or mailing 
lists. Much of the best feedback i get is from looking for other forums 
where people are discussing my software. They are more likely to discuss in 
their existing communites than join new ones.

13) Many Windows users are intimidated by linux dominated communities. True 
or not, the perception is that most linux/open-source advocates are grouchy 
one-eyed zealots.

As the joke goes the best way to get information from a linux dominated 
community is not to ask "How do i do xyz" and be greeted by "Read the 
source", "Read the man page", "Google is your friend", "Don't waste my time 
you should know to dingle the doohickey before dangling the doowop" or "What 
are you doing you noob, everyone know to post doohickey related issues in 
the wubblewub list"

In order to get the most information from such a community, you have to go 
in and say "abc is hopeless because it can't do xyz" at which point you'll 
be bombed with 100 different ways to do xyz.

Whether that's the reality is relatively unimportant, that's the perception.

14) Users come first, in general the developers will follow the users.

Probably 25% of people who've downloaded my directshow filters have never 
even seen my webpage or xiph's. (They get there from third party download 
sites which link directly to binaries) Now that sounds like a bad thing... 
and in some ways it can be, it's hard to connect to users when they don't 
know who you are.

But the other way to look at it is, that users generate interest and 
interest generates more users. Until you get a critical mass of interested 
users demanding more and better applictions for their platform, it will 
always be hard to find developers to write those applications.

If "losing contact" with 25% of the users for a while is the price you have 
to pay to just get the software out on more machines... so be it. Eventually 
most of those will "regain contact"... some never will... that's just the 
way it goes. And ideally that lost 25% helps to raise awareness and increase 
demand which in the long run is beneficial.

Once you have a substantial userbase then you can worry about creating a 
self-sustaining community for that platform because you will have enough 
interested users and developers to make it successful.

Anyway... that's just my thoughts... conratulations to anyone who actually 
read all those ramblings :)

Cheers,

Zen.